BCCA Cradle Project - C123

 

BCCA Cradle Project
Case study
Document Type	Case study
Document Identifier	123
Objective functions	Cost Maintain Rate Increase Quality Increase
MSTE workflow	Development
Prerequisites

Challenge

The CARA cradle is a patient positioning system used during cancer treatment, developed through a collaboration between the Composites Research Network at the University of British Columbia and the British Columbia Cancer Agency (BCCA). Early prototypes confirmed that a composite cradle was feasible, but they also revealed significant design and manufacturability challenges. Issues such as foam cracking during thermoforming, difficulties with prepreg layup, and excessive labour requirements limited consistency and scalability were identified. The goal across this project became to iteratively refine the cradle’s design to achieve better clinical fit, stronger structural performance, and a repeatable, cost-effective manufacturing process.

Video

Approach

The project evolved over multiple design and manufacturing phases following the MSTEP approach:

Phase 1

The primary objective of Phase 1 was to develop a baseline prototype that could serve as a reference for future iterations and suitable for use in clinical trials.

Materials Selection

The most significant change in the first iteration was the switch from a monolithic carbon fibre composite part to a sandwich panel construction. To enhance the radio transparency of the cradle and reduce weight, thin-ply NTPT prepreg was applied over a Rohacell 31HF foam core. The cradle was formed using a two-piece machined foam tool, which is bonded together after the layup process. This sandwich construction provided increased stiffness while lowering overall density. The driving factors for material selection were high relative strength, low density, bond quality between the core and fibres, and fibre formability. The choice of core material and prepreg thickness was based on finite element analysis (shown in Figure 1) and mechanical testing (shown in Figure 2) to ensure the cradle could safely withstand anticipated load. Various layup schedules, fibre types, core materials, and thicknesses were evaluated, ultimately selecting 2 mm Rohacell 31HF as the core with custom thin-ply prepreg as the skin. Two separate prepreg rolls were used, with the tool-side material incorporating an additional resin film layer to improve surface finish.

Figure 1 - Simulation of deformation under uniform pressure

Geometry Simplification

The pyramid mounting section (used to hang the cradle in Figure 2) quickly emerged as a major manufacturing challenge. Mechanical testing revealed stress concentrations at the transition between the pyramid section and the rest of the cradle.

File:Cradle in a highly deformed state qqDHInQApCp4.png

Figure 2 - Cradle in a highly deformed state

Its complex curvature made fibre placement and cutting difficult, even with the use of templates, and the process was time-consuming and hard to scale for higher production volumes. To accommodate the curvature, darts were added to the fabric to reduce wrinkling, but this often led to inconsistent composite coverage and foam defects. Forming the foam over this geometry was also challenging, creating additional imperfections. The complex geometry of the initial prototype required a specialized vacuum bag with custom inserts to maintain adequate compression. While a well-defined process allowed these bags to be produced consistently without leaks, it was labor-intensive and still prone to defects such as wrinkling, bridging, and incomplete consolidation around the pyramid. Additional simulations in Abaqus FEA software (Figure 3) were performed to evaluate alternative geometries and ensure the structural performance of the revised design matched the original pyramid mounting feature.

Figure 3 - Abaqus simulation of pyramid mounting design vs double curvature design

Cure Cycle

The next step involved analyzing the cure cycle to ensure it was optimized and aligned with the manufacturer’s recommended cure cycle. Heat surveys were conducted by placing thermocouples throughout the part, tooling, and oven. These measurements revealed that the part’s large thermal mass prevented it from reaching the desired cure temperature, even with extended temperature holds. This led to a redesign of the tooling to reduce thermal mass. During curing, it was also observed that the prepreg shifted over the foam core, resulting in suboptimal positioning over the foam core. This issue was resolved by shortening the first temperature hold, minimizing the time the resin remained at low viscosity and reducing prepreg movement.

Defects of Phase 1

While the process produced functional parts, several challenges remained. The geometry had been improved but still included complex double curvature, which complicated foam forming and limited scalability for higher-volume production. Additionally, finishing the edges proved labor-intensive and inconsistent, making it unsuitable for routine manufacturing.

Phase 2

Phase 2 focused on refining tool geometry, the layup process, and foam forming, though the remaining double-curvature design continued to present challenges for both foam formability and prepreg drapability.

Foam Forming

Producing the foam section required testing the cutting of the foam and the formability over both the L-shaped and pyramid sections (before they were removed). Rohacell foam proved particularly tricky to form, as it needed to be precisely at its glass transition temperature (Tg) to bend without snapping. The manufacturer’s recommended approach, placing the foam and tool in the oven to reach thermal equilibrium at Tg, then forming it quickly, proved difficult in practice. When the oven doors were opened, the foam temperature dropped too quickly, causing snapping during forming. Attempts to compensate by reheating the foam between bends had limited success.

Figure 4 - Pyramid foam forming test

The most effective method involved heating the oven above Tg and placing the tool, foam, and thermocouple inside before the oven reached the target temperature. A thin rubber sheet was placed on top to hold the foam edges down, and a small metal weight applied over the curved region. The system was allowed to equilibrate until the thermocouple reached Tg, after which the tool was quickly removed, and the foam bent to shape and held until cooled to 80 °C. This approach successfully produced the part shown in Figure 5. Despite this success, the complex geometry of the tooling continued to cause defects in the foam, including unevenness, wrinkling, cracking, and incomplete translation of the curvature into the final composite part, highlighting the ongoing challenges of producing high-quality components from such intricate designs.

Figure 5 - L-section foam forming test

Several techniques were explored to form the foam, including pre-shaping with a heat gun, applying vacuum in the oven, and using a silicone male tool. While some methods showed promise, none produced a consistent thermoforming process. Once formed, the foam needed precise cutting to remove excess material, as shown in Figure 5. A 3D-printed tool and razor blade were used to achieve a clean surface finish.

Layup

Building on Phase 1 simulations, the cradle was designed as a sandwich panel, with low-density foam at the core and unidirectional prepreg on the outside. Layup schedules were chosen based on simulations and mechanical testing to maximize strength while minimizing weight. Initially, a thinner foam core met density requirements, but increasing the thickness to 5 mm improved stiffness and allowed for carbon removal around the edges. A monolithic mounting section was also added to strengthen the attachment area and improve quality around drilled holes. During prototyping, challenges arose with prepreg sliding on the mold and difficulty forming it over the foam. To limit the movement of the fibre during the layup, a spray adhesive was used to keep it in place. Twelve prototypes were produced to test different wrapping techniques and surface finish improvements. While layup time could be reduced, edge finish remained a challenge, leading to the decision to increase foam thickness to simplify the design.

Monolithic Mounting Section

To further improve strength and eliminate potential failure points, the mounting section was switched to monolithic carbon fiber. This change added a non-crushable area for attaching the mounting fixture. As this area lies outside the radiation zone, density reduction was not critical. The change improved part quality, preventing exposed foam around drilled holes and ensuring the carbon finish met sanitation standards. Finally, joining the monolithic section with the sandwich panel required design optimization. Initially, the carbon layer faced the patient side with foam underneath, but this was later reversed, as shown in Figure 6. Placing the stronger carbon material on the underside of the curve improved part strength, as this region experiences compression, reducing the risk of buckling and prepreg surface wrinkling.

Figure 6 - Taper design of monolithic section

Phase 3

A structured effort was made to redesign the cradle for manufacturability following the MSTEP approach.

Geometry Optimization

Over the course of the project, the geometry of the cradle evolved significantly. Figure 7 illustrates this progression, from the first version with a pyramid mounting section, to the second with complex double-curvature, and finally to the third, which could be reliably produced in large quantities. As the design matured, it became simpler. This is a common theme in prototyping and design iteration.

Figure 7 - Geometry progression

The most substantial geometric changes occurred during Phase 3, when ten different part geometries were tested to optimize manufacturability, radiation zone clearance, and support placement. Adjustments included altering curvature angles to position supports further from the mounting section and reducing the size of the mounting area. The final design, shown in Figure 8, is fully symmetric and represents the simplest and most practical geometry.

Figure 8 - Model 10 Prototype (without edge finishing)

This simplified shape eliminated the need for cross-sectional changes other than rounding of the corners, allowing higher-volume production from larger sheets. Achieving this simplification required several considerations: the layup had to withstand expected forces without the pyramid mounting section, the monolithic edge needed sufficient strength to prevent buckling, and mounting holes were repositioned to minimize stress concentrations at the foam-carbon transition. The new shape of the part required new tooling to be machined. With this new tool design, larger panels could be made and cut into multiple parts, with the limitation being the size of the oven. This approach required rethinking edge processing, as layup could no longer fully encapsulate the edges, prompting the development of improved edge-finishing techniques.

Edge Finishing

Edge finishing is critical for both patient safety and sanitation. Sharp edges can cause injury, and exposed foam is porous and difficult to clean. Initially, edges were covered with a monolithic carbon section from the layup, but this approach produced sharp, labor-intensive edges and occasional exposed foam. Flat-panel tests explored alternative geometries and filler materials, which provided smooth finishes but failed to meet radio-transparency requirements. The option of wrapping the entire cradle with shrink-tubing was explored but quickly abandoned as this method resulted in heavy wrinkling around curved sections and was unable to accommodate any changes in cross-section. Ultimately, it was found that vinyl-wrapping the edge and coating it with Cerakote paint produced the most desirable result. The vinyl was effective for covering any sharp edges produced by the carbon fibre, and the Cerakote provided a good seal to make the transition of carbon to vinyl more robust and easier to sanitize. This method also yielded an aesthetically pleasing finish, making it ideal for commercialization.

Tooling Innovation

With an improved edge finishing technique and simplified part geometry, more opportunities for tooling improvement became clear. As the edges no longer needed to be covered during the layup, it became possible to shift the design for higher-volume production. As seen in Figure 9, a single continuous layup could now produce multiple parts with simple and limited post-processing. Reducing the number of layups,the most labor, and time-intensive step, greatly improved overall manufacturability.

Figure 9 - Parts made from a single tool and layup

The new single-curvature tool simplified and accelerated the layup process, as the prepreg no longer had to navigate complex curves or endure excessive shear stress. This reduced wrinkling and other layup defects, improving overall part quality. Post-processing was also enhanced, with dedicated tools and fixtures developed to consistently cut cradles to precise dimensions and accurately drill mounting holes in the monolithic section. With these improvements, the cradle design became suitable for clinical trials and higher-volume production.

Figure 10 - Cradles manufactured for clinical trials prior to wrapping and painting

Outcomes

• Manufacturability Improvements: Simplified geometry and wide-panel layup reduced production time and improved part consistency.
• Final Geometry: The final model adopted for clinical use is symmetrical, less curved, and easier to form.
• Edge Safety: Vinyl wrapping replaced labour-intensive monolithic edge layups.
• Repeatability: New cutting jigs and manufacturing guides supported consistent cradle trimming and drilling.

Impact

Across three phases, CRN’s iterative design-build-test workflow significantly advanced the clinical readiness of the CARA cradle. The evolution from early, highly curved prototypes to streamlined, manufacturable designs illustrate the value of integrating design for manufacturability from the outset. These learnings support future scale-up and potential production automation for composite medical devices.

Key Techniques

• Thin-ply prepreg layup (NTPT CF1) on thermoformed Rohacell foam
• Geometry refinement via CAD-based design iteration
• Vacuum bagging with semi-rigid rail-based thermoforming setup
• Prototype validation via structural testing and clinical feedback
• Wide-panel layup and cutout strategy for efficiency and edge quality

About	Help

Engineered materials (designed to have specific properties) made from two or more constituent materials with different physical or chemical properties. The constituents remain separate and distinct on a macroscopic level within the finished structure.

Pre-impregnated (prepreg) material refers to fibre that is already combined with resin. It is the most common material form used in aerospace.

During prepreg production, (e.g. fibres are run through a resin bath), prepreg is heated and partially cured to B Stage (< 5 % degree of cure). Thermoset prepregs (e.g. epoxy prepreg) have to be kept in a freezer at around -20 °C. At room temperature, the epoxy starts to cure.

Material and process (M), Shape (S), Tooling and consumables (T) & Equipment (E) - all have an interlinked effect on the Process step (P). (See MSTE factory ontology)

Carbon fibres are composed of large aromatic sheets similar to those in graphite. These graphitic layers form the basic structural units in the shape of ribbons. The structure of carbon fibre ribbon is believed to be a columnar arrangement of disoriented graphite crystallites parallel to the ribbon length. The idealized tetragonal crystallites are stacked above one another, with slight disorientation between the crystals in the direction of fibre axis, trapping sharp needle like voids, where the boundaries between the stacks represent the disordered regions.

For polymer matrix composites (PMCs), resin refers to the matrix; the continuous material phase that binds the reinforcement together, maintains shape, and transfers load. Resins are divided into two main groups: thermosets and thermoplastics.

In composites processing, viscosity is an indicator of how easily the resin matrix will mix with the reinforcement and how well it will stay in place during processing. The lower the viscosity, the more easily resin flows. Resin viscosity ranges considerably across chemistries and formulations.

By scientific definition, viscosity is a measure of a material’s resistance to deformation. For liquids, it is in response to imposed shear stresses.

The glass transition temperature (T_g) is the temperature region where the polymer transitions from a hard, glassy material to a soft, rubbery material. It is one of the most important properties of any amorphous polymer.

Any manufacturing and/or decision making activity that occurs during any stage of the development design cycle (e.g. conceptual design to production).

In the context of Knowledge in Practice, practice refers to the systematic use of science based knowledge to reduce composites manufacturing risk, cost, and development time.

With regards to manufacturing, risk is the combination of the probability and consequences of undesirable manufacturing outcomes. Manufacturing risk can lead to technical issues, program/schedule delays, and cost overruns.

A key component of all composite manufacturing processes. Collectively, the four themes represent the time-temperature-pressure-vacuum history, which is traditionally used to define a manufacturing cycle.

The four processing themes are:

• Thermal management
• Material deposition management
• Flow and consolidation management
• Residual stress and dimensional control management

(Same as "Processing themes")

With regards to modelling & simulation, validation demonstrates the accuracy of a computational model — in the context of its intended use — to check that it represents the physics of the problem and the ‘reality of interest’ (needs are satisfied).